Doctoral Degrees (Food Science)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/537

Browse

Browsing Doctoral Degrees (Food Science) by Author "Du Preez, Brigitte Von Pressentin"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Development of a quality grading system for the honeybush (Cyclopia spp.) tea industry
    (Stellenbosch : Stellenbosch University, 2020-12) Du Preez, Brigitte Von Pressentin; Joubert, Elizabeth; Muller, Magdalena; Moelich, Erika Ilette; Stellenbosch University. Faculty of AgriSciences. Dept. of Food Science.
    ENGLISH ABSTRACT: Honeybush tea, produced from Cyclopia species endemic to South Africa, has attained an international footprint within the global herbal tea sector. As demand is exceeding supply to a primarily export market, all production batches should meet optimum quality standards. The lack of standardised sensory quality criteria and assessment methods within the commercial sector has resulted in tea of variable sensory quality reaching the market. The trade of inconsistent and inferior quality products will be detrimental to the reputation of honeybush tea and consumer acceptance, and ultimately the honeybush industry. The need for a scientifically founded quality grading system to evaluate, differentiate and communicate the sensory quality of honeybush tea was addressed through four quality control elements, i.e. a sensory lexicon and wheel, sensory quality standard, quality scoring method, and rapid quality classification methods. The previously developed honeybush aroma lexicon and wheel were revised, based on a newly established comprehensive sensory and physicochemical dataset. Data of samples of the main commercial Cyclopia species (C. intermedia, C. subternata and C. genistoides), processed on laboratory- and commercial-scale, were incorporated to represent the sensory space in terms of different qualities. Universal chemical-based reference standards were developed and validated to replace food-based reference standards in the aroma lexicon to facilitate standardised assessment of honeybush sensory quality. The established sensory quality standard was founded on the comprehensive dataset and input from industry. Sensory quality parameters for the tea infusions and dried plant material were identified, and parameter specifications for ‘high’, ‘moderate’, ‘low’ and ‘poor’ quality classes were defined through expert focus groups. A user-friendly quality scoring method that incorporates a scorecard and colour reference card, was developed and validated for the assessment and classification of production batches based on obtained parameter and total score values and citation frequencies of specific attributes. The validity of reference-based rapid methods, polarised sensory positioning (PSP) and polarised projective mapping (PPM), were investigated for their discrimination ability as time-efficient classification tools to distinguish between infusions of large samples sets of variable sensory quality within commercial and research context. The efficacy of the use of physical (p) poles (tea infusions) and novel theoretical (t) poles (descriptions), representative of the four sensory quality classes, as references, were compared within each method, using a trained panel. Product configurations similar to that of a classical sensory profiling method, descriptive sensory analysis, demonstrated the validity of the method variations for broad quality classification based on key sensory quality parameters. PPM-p indicated the highest discrimination ability between the quality classes. Recommended amendments to theoretical pole descriptions would improve feasibility for commercial application. The quality scoring method and PPM-t were tested by a panel of industry representatives, and the need for industry assessor training in sensory quality parameters was emphasised. Implementation of the proposed integrated quality grading system will equip honeybush industry role-players in delivering a final product of consistent sensory quality within the honeybush value chain.